334 



of the gaDglion, the other of large cells like the large cells present 

 in the ganglion in the zone of origin of the nerves. In the chain 

 form of S. pinnata I was able to show that there were nerve fibres 

 connecting the larger-celled outgrowth with the ganglion itself. 



In the chain form of S. cordiformis I have recently demonstrated 

 a similar connection between the ventro-lateral outgrowths and the 

 ganglion (Fig. 4). In this species there is a single pair of large 

 masses of cells, lying on each side of the ventral part of the ganglion. 

 They are somewhat sickle-shaped, being convex toward the ganglion. 

 The cells of which they are composed are large and irregular in 

 shape (Fig. 5). Their protoplasm is clear, and they have irregularly 

 thickened walls, which stain deeply with hsematoxylin. The whole 

 histological appearance of these organs in section is very closely 

 similar to that of the dorsal eye in the chain form of this species, or 

 in the chain S. hexagona, or of the smalller eye in the chain S. run- 

 cinata. Organs of the same character are found in the same position 

 in the solitary S. cordiformis, but their innervation is difficult to 

 demonstrate. 



This adds another to the strange series of conditions we see in 

 the Salpas, showing modification of nerve cells into rod-like cells. It 

 would seem absurd to call such cells, as those here described in 

 S. cordiformis, rod cells, were it not for the complete series of gra- 

 dations between these irregular cells and such perfect rods as we 

 find in the small eyes of S. pinnata (Fig. 6). When we remember 

 that in the large dorsal eye of the chain S. hexagona or S. costata 

 (Fig. 7) we see all the rod cells of this same character, and that in 

 that species they are undoubtedly visual, being associated with well 

 developed pigment cells, it suggests that these irregular cells in the 

 outgrowths from the ganglion in S. cordiformis may be truly rod cells. 



In the Salpas, in addition to the large dorsal eye, so long known, 

 I have described smaller eyes or masses of rod cells, which, in the 

 chain individuals of different species, may occupy the following posi- 

 tions — the posterior mid-point of the ganglion (paired); the dorsal 

 surface of the ganglion, at the base of the larger eye (paired or un- 

 paired) ; the postero-ventral area of the ganglion (unpaired) ; outgrowths 

 from the mid-lateral areas of the ganglion, outgrowths from the ventro- 

 lateral areas of the ganglion. There seems to be in the chain Salpae 

 a general tendency to the formation of rod cells at different points 

 in the ganglion or its outgrowths. Pigment may or may not be found 

 in connection with these masses of rod cells. 



Are these so-called eyes functional as optic organs? 



